Skip to content
Company
Management teamCareersNewsLocationsPartners
BlogCustomer Portal
  1. HomeArrow Icon
  2. Case StudiesArrow Icon
  3. How Thermo Fisher Scientific transformed test automation with Tricentis Tosca and Data IntegrityArrow Icon

Thermo Fisher Scientific

How Thermo Fisher Scientific transformed test automation with Tricentis Tosca and Data Integrity

%+

savings of regression testing time

default

K+

tests automated with Tosca

default

Company overview

Thermo Fisher Scientific Inc. is a leading biotechnology and laboratory supplies company, with annual revenue of approximately $40 billion. Its mission is to enable its customers to make the world healthier, cleaner, and safer, whether it’s advancing life sciences research, solving complex analytical challenges, increasing productivity in laboratories, improving patient health through diagnostics, or developing and manufacturing life-changing therapies.

Thermo Fisher Scientific delivers a combination of innovative technologies, purchasing convenience, and pharmaceutical services through its industry-leading brands, including Thermo Scientific, Applied Biosystems, Invitrogen, Fisher Scientific, Unity Lab Services, Patheon, and PPD.

Challenges

  • Ensuring no defects in a regulated environment
  • Finding a more readable report format for stakeholders
  • Too many testing tools for different domains

Simplifying test automation tools in a regulated environment

Gaurav Mittal works as a QA manager for the data science domain at Thermo Fisher Scientific.

“Software lies in the heart of Thermo Fisher, and when we have to provide technical solutions, we want to make sure the data is accurate,” Mittal said. The team must ensure data analysis is correct, production deliveries are robust, and there are no defects in the scope — especially in a regulated environment.

One of the key goals for Thermo Fisher Scientific was to enhance its test automation coverage. Recently, the organization moved to using GitHub Actions as a CI/CD tool. With GitHub Actions, in one click the team could execute the automation test. However, it was only generating reports in XML format. XML format was not readable by its business stakeholders, and they found it challenging to figure out.

When Mittal first joined Thermo Fisher Scientific, the organization was using many different tools and technologies, such as Postman for API testing and Selenium for scripting language. The company needed one tool that could talk to all the different domains, like GitHub Actions, and be easily integrated.

Tricentis Tosca automates data quality, integrity, and validation

Thermo Fisher Scientific selected Tricentis Tosca as its platform for automation. Mittal said that Tosca can easily integrate with Thermo Fisher Scientific’s domains like ERP, web, and data.

“It’s a big win, and now when we are executing our regression tests which are in like more than thousands of the count, we are able to save sixty to seventy percent of the manual execution time frame,” Mittal said.

Mittal thinks automation is necessary for the data validation checks. “There are several tables and within one table only, per day, we are seeing thousands of rows. They are being entered, and we need to make sure the data is not corrupted,” he noted. “It is accurate and when we deploy the code in production, we are not going to face any hurdles, like no back and forth. We don’t want that. It consumes a lot of time.”

Tosca also helps to ensure quality and data integrity for Thermo Fisher Scientific’s regulated environment. The regression test suite is built as part of its quality gate and gives Mittal’s team the confidence before deploying the code into production and ensures the existing functionality will not break.

“When it comes to data integrity practices, we are using Amazon databases like Redshift and when the customer is processing data through the web application or any other form, data can be corrupted,” Mittal said. “So we need at our end a quality gate when we are building. We need to make sure the validation checks which involves like null check, mandatory fields, referential integrity, all these, they are working absolutely fine.”

Vision AI boosts further efficiency in test case generation

“Our mission is to be innovative and to come with digital cutting-edge technology solutions,” Mittal said. “And as part of this innovation, my vision was for the web applications, how can we utilize AI and as well as how can QA get more visibility because unless code is deployed, QA has no role in that part.”

Mittal is overcoming both of these gaps with Tricentis Vision AI. With the tool, QA can do whiteboard diagrams and mockups, and “is getting a chance to be at the front seat and generating the test cases.” Moving forward, the team is focused on using Vision AI and predictive analysis to figure out patterns and reasons why tests might fail.

“Now with the Vision AI, you can feed mockups, whiteboard diagrams, and Vision AI generates the test cases so that QA can execute as soon as the UI is being developed and deployed,” Mittal shared. “It’s like you have moved testing shift left and now at the very beginning of the stage, like before the planning, QA is also getting visibility, and their work is also getting started. So, it’s a big win by Tosca tool.”

Mittal also cited Tricentis support for its premium services, technical abilities, and getting back to the team within a 24-hour time frame.

“They know each and every depth of the feature…It doesn’t take long to make them understand where we are stuck, what we can do.”

Streamlining test automation with one click

“Now, Tricentis Tosca is our major platform for automation, and we have written like thousands of automation tests,” Mittal said.

With Tosca, Mittal’s team found a way to generate reports with GitHub Actions in PDF format. It’s a more readable and less technical format versus XML, so stakeholders can easily figure out what is failing.

“Now with Tosca, once you have developed your test suite, with one click, you are executing thousands of tests together, and a PDF-generated email will be sitting in your inbox,” Mittal said. “On top of the report, it shows a pie chart on how many are passing, how many are failing. So that also gives a fair idea on the confidence level, whether you can deploy it or not.”

Mittal also likes the robustness of the Tosca locators and the benefit of not having to rework a script repeatedly. “So that is one major win where automation resources face a lot of challenges. A lot of maintenance activities are required, but with the Tosca scripts, we are not facing that much of effort in the maintenance area.”

Results

  • Using Tosca resulted in 60% to 70% savings in regression testing time versus manual execution
  • Use of Vision AI has increased efficiency and shifted testing left
  • Thermo Fisher Scientific can now generate reports with GitHub Actions in a more readable PDF format versus XML

Get started with Tricentis

Get started

Arrow